Computational Statistics

点击翻译按钮获取中文摘要

Today, increasingly larger and more complex data are being produced. Ultra high-dimensional data are among the most problematic. Statisticians and computer scientists can deal with high-dimensional data, which can contain thousands of variables. Data being produced today, however, can have tens of thousands of variables and even more. Unfortunately, methods that work for high-dimensional data are often unsuitable for ultra high-dimensional data. Computational statistics approaches for ultra high-dimensional data will be developed, focusing on methods that find homogeneous subgroups or clusters. These approaches will be applicable in any setting where ultra high-dimensional data arise, from management science to disease diagnostics and bioinformatics.